Aves. — In the bird the ganglia present much the same structure, both as regards the ganglionic corpuscles and nerve-fibres, as those of the mammalia.
— In the frog, the animal in this class which has been most frequently ex amined, the ganglionic corpuscles present the same general characters as those in the higher animals. The existence of apolar and uni polar ganglionic corpuscles is much better seen in examining the small ganglia in the heart or bladder of this animal than in the ganglia of the mammal or bird. Bipolar cells have been described by Schiff; Valentinf has also de scribed and figured bipolar ganglionic cor puscles the nerve-tubes of which ran in opposite directions, one towards the centre, the other towards the periphery. They were found by him both in the small ganglia of the heart, and also in the ganglia occurring in the main chain of the sympathetic. The nerve tube connected with these he describes as clear and transparent, differing from the broad nerve-fibres in its colour and general appear ance, especially in its not presenting any oily contents, and thus appearing to belong to the fine variety of fibres. Kolliker describes the unipolar ganglionic corpuscles as being pyriform, and at their narrow extremity drawn out into a delicate process. This presents the same pale and finely-granular appearance as the corpuscle itself, and measures from the A-,,--6th to the of an inch in dia meter : after running a short distance from the ganglionic corpuscle it acquires a dark margin and slightly granular contents ; becorning, in short, a fine nerve-fibre (see fig. 289. A). Bidder has also observed in the ganglia of the heart of the frog bipolar cells, which, however, resembled the unipolar in the fact that both nerve-tubes ran in the direction of the periphery. In the ganglia of the frog there is a much smaller amount of the substance present, which has been described as consti tuting the capsules of the ganglionic corpuscle, and there are also very few of the fibres of Remals.
Pisces. —In certain animals belonging to this class, especially in the cartilaginous fishes, the :connection between the ganglionic cor puscles and the nerve-tubes is much better seen than in any of the preceding classes. In the torpedo and ray- the bipolar variety- of ganglionic corpuscles was first discovered by Wagner, and shortly afterwards by Robin. Similar observations were also made by Bid der and Reichert, both as regards the car tilaginous fishes, and, by the latter observer, in the cod, perch, and certain species of sal mon, as well as in the pike.* In the common ray the ganglionic corpuscles, as occurring in the spinal ganglia, are generally more or less round or oval, and are much larger than in any of the higher animals. They- measure from 4L,th to the 1 oth of an inch in dia meter, and contain a more or less clear viscid fluid with finely molecular matter. On the addition of diluted acids or spirit, they be come dark and granular. Each of the cells contains a clear round nucleus, in which there is also present one, sometimes two, nucleoli.
The contents of the nucleus, like those of the cell itself, become dark and granular on add ing the reagents above mentioned. In se veral of the cells there are seen, apparently on the inner surface of their wall, a num ber of round corpuscles, generally clear and transparent, but sometimes more or less dark and granular. They measure about the llooth of an inch in diameter, and seem, as Wagner and Robin describe, to form a single layer on the inner surface of the ganglionic corpuscles. The larger of the ganglionic corpuscles are generally more or less spheri cal ; the smaller, on the other hand, present commonly a more oval shape. Sometimes between the outer cell-wall and the contents of the vesicles there is, as Bidder describes, a clear space, varying in breadth, generally broadest at the points where the two nerve tubes are connected with the corpuscle ; at other times this space does not appear to exist, the granular contents of the vesicle coming close up to the cell-wall (see A. and D.fig. 286.). The wall of the corpuscle appears to be much stronger than that of those in the higher animals, and is distinctly prolonged at either extremity into the mem brane of the nerve-tube, the mo constituting one continuous structure, agreeing both in anatomical characters and in their relation towards reagents. When the ganglion-ve sicle is ruptured, so as to allow its con tents to escape, its cell-wall collapses more or less, and often presents the appearance of lines passing outwards towards the circum ference, from a central point (tr. fig. 286.) • this appearance is evidently due to folds in the membrane. In the clear space which has been mentioned as sometimes existing be tween the cell-wall and the contents of the ganglionic corpuscle, there are often observed a number of particles, most abundant at the two poles of the vesicle, which resemble in appearance the curd-like contents of the nerve tubes, and are evidently continuous with the same (c, fig. 286.). Sometimes the granular contents of the vesicle appear to be prolonged downwards into the nerve-tube (A. fig. 289.), appear on addition of acetic acid. In other parts the ganglia contain corpuscles similar to those already described as occurring in the spinal ganglia, with the exception that they are in general perhaps somewhat smaller. They are imbedded in a fibrous structure, which seems to hold the same relation to them as the nucleated substance forming the capsules of the ganglionic corpuscles in the higher animals. It is difficult to ascertain how far the ganglionic corpuscles in the ab dominal ganglia are unipolar or bipolar; ac cording to Wagner, they are the same in this respect as the corpuscles in the spinal ganglia, being all bipolar ; one tube entering while another leaves the corpuscle. The nerve tubes which are connected with them belong both to the broad and fine varieties ; in ge neral, the narrower fibres are connected with the smaller corpuscles, the broader fibres N it h those of larger size (fig. 290.).